AI System for Routing Transport Damage Images to a Review Queue

The problem

In day-to-day operations, a damage image may not be linked to the correct load and stage. When damage photographs, package identifiers, capture times, transfers and delivery records are scattered across sensors, documents, field records or team files, issues may not become visible in time.

This fragmented data delays preparation of the damage review file and forces teams to search manually for supporting evidence again. The logistics damage specialist must complete missing records and verify each conflict before making a decision.

A common exception is that a mark on the packaging may not indicate damage to the product. Incomplete or incorrect automation that fails to check this context could attribute liability incorrectly or trigger an unnecessary compensation process, so no physical or operational action can be carried out directly.

What we set out to improve

• Proportion of review candidates verified by the damage specialist
• Proportion of outputs corrected or rejected by the logistics damage specialist
• Time between the source event and the review output

How the system works

1. 01 Define the review scope and the role of the authorised logistics damage specialist
2. 02 Retrieve the inputs: damage photographs, package identifiers, capture times, transfers and delivery records
3. 03 Check source identity, date, calibration or version, and data recency
4. 04 Match each image to its package and capture time
5. 05 Perform an initial classification of visible damage and missing evidence
6. 06 Check for exceptions: a mark on the packaging may not indicate damage to the product
7. 07 Place low-confidence or conflicting records in a separate specialist queue
8. 08 Write the source identity, rule, output and timestamp to the audit trail
9. 09 Have the logistics damage specialist approve, correct, defer or reject the output

Methods we used

• Initial image classification and delivery-chain matching
• Source, format, timing and required-field checks for damage photographs, package identifiers, capture times, transfers and delivery records
• A missing-data warning instead of an estimate when inputs are incomplete or conflicting
• A separate exception check outside the main rule
• Combined retention of the source, rule result, damage review file and human decision

Where people stay involved

Data and security

Access to damage photographs, package identifiers, capture times, transfers and delivery records is restricted to the minimum permissions required for the task. Commercial, location and employee information is masked where possible, and only the necessary portion of the dataset is sent to the model. Access, outputs and the logistics damage specialist's decision are logged; production use does not begin until the organisation has approved the retention period and data location.

Who this suits

Frequently asked questions